Is Weight Tying Still Beneficial for Decoder-Only LLMs in Private Settings Under DP-SGD?
Research
Research News
New papers, breakthroughs and academic AI research
The System Prompt Illusion: How Instruction Preambles Modify Computation in Language Models
Research
Make Code as Policy Great Again: Frontier Agents Write, Call, and Evolve Robot Tools
Research
Removing Timing Shortcuts Improves Non-Invasive Brain-to-Text
Research
DynaFlow: Transparent and Flexible Intra-Device Parallelism via Programmable Operator Scheduling
Research
Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout
Research
Trust Is Not a Score: Runtime Assurance Contracts for High-Risk AI Agents
Research
OSWorld-Science: A Benchmark of Computer Use Agents for Learning and Using Scientific Software
Research
OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong Adaptation
Research
CombEval: A Framework for Evaluating Combinatorial Counting in Large Language Models
Research
NeuroAI and Beyond: Bridging Between Advances in Neuroscience and Artificial Intelligence
Research
Safety of Latent Communication in Multi-Agent Systems
Research